THE JOURNAL OF BIOWGICAL CHEMISTRY Vol. 269, No. 19, Issue ofMay 13, pp, 14021-14026, 1994 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

Prolactin-induced Proliferation of Nb2 Cells Involves Tyrosine Phosphorylation of the Prolactin Receptor and Its Associated Tyrosine Kinase JAK2*

(Received for publication, November 29, 1993, and in revised form, February 15, 1994)

Jean-Jacques LebrunS, Suhad AliSP, Luc SoferS, Axel UllrichO, and Paul A. Kellyln From the VNSERM Unite 344, Endocrinologie MolQculaire, FacultP de Medecine Necker; 156 rue de Vaugirard, 75730 Paris, France and the §Department of Molecular Biology, Max-Planck Znstitute fur Biochemie, Am Klopferspitz 18A, 8033 Martinsried, Germany

The interaction of prolactin with its receptor in the Nb2 cell line has been shown to induce the phospho- rylation of cell-associated proteins and mitogenesis. It has been reported previously that one of these proteins, phosphorylated upon prolactin stimulation, was a tyro- sine kinase. We have identified this kinase as JAK2, and demonstrate its association with the prolactin receptor. In addition, we show that the prolactin receptor itself becomes tyrosine phosphorylated upon ligand stimula- tion in Nb2 cells. These actions are time-dependent and occur rapidly after prolactin stimulation, with first the kinase being activated within 5 min and then the recep- tor being phosphorylated maximally at 20 min. More- over, phosphorylation of both JAK2 and the receptor as well as Nb2 cell proliferation are dependent on the con- centration of lactogenic hormone, resulting in a bell- shaped response curve similar to that observed in the two site model of hGH action. This indicates that early events in signal transduction as well as later events like mitogenesis and proliferation involve prolactin recep- tor dimerization. Together these data indicate that the prolactin receptor in Nb2 cells is associated to JAK2 and that upon ligand stimulation, and receptor dimeriza- tion, the kinase and the receptor are tyrosine-phospho- rylated, which represents the first event in the process of prolactin receptor signal transduction in Nb2 cells.

In the complex new biological field known as the neuroim- munoendocrine system (Comsa et al., 1982), PRL’ appears to play a critical role, by modulating lymphocyte functions by endocrine, paracrine, or autocrine interactions. PRL itself is present in T cells (Russell et al., 1984; Pellegrini et al., 1992; Sabhanval et al., 19921, and its receptor appears to be ex- pressed ubiquitously in all the immune cell types (Pellegrini et al . , 1992).

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed: INSERM Unite 344-

Endocrinologie Moleculaire, Faculte de Medecine Necker, 156 rue de Vaugirard, 75730 Pans, France. Tel.: 331-4061-5310; Fax: 331-4306- 0443.

‘The abbreviations used are: PRL, prolactin; bGH, bovine growth hormone; DSP, dithiobis(succinimidy1propionate); ECL, enhanced chemiluminescence; EPOR, erythropoietin receptor; GH, growth hor- mone; oGH, ovine growth hormone; hGH, human growth hormone; GHR, growth hormone receptor; gp130, glycoprotein 130; JAKl, JAK2, just another kinase 1 and 2; oPRL, ovine prolactin; hPRL, human pro- lactin; PRLR, prolactin receptor; Nb2, rat lymphoma cell line; PAGE, polyacrylamide gel electrophoresis; PME, P-mercaptoethanol; mAb, monoclonal antibody.

Molecular characterization of the PRLR (for review, see Kelly (1993)) indicates that the receptor belongs to the newly de- scribed cytokine/GH/PRL receptor family (for review, see Cos- man (1993)). The intracellular signaling pathways for the different members of this family have not been fully character- ized.

In the rat, two different forms of the PRLR, which differ mainly in the length of their cytoplasmic domains, have been identified: a short form of 291 amino acids and a long form of 591 amino acids (Boutin et al., 1988; Shirota et al., 1990). A mutant form of the PRLR, caused by a genetic deletion result- ing in a cytoplasmic domain shortened by 198 amino acids, has been identified in the Nb2 rat lymphoma cell line (Ali et al., 1991). This cell line was derived from a lymph node tumor of an estrogen-treated male rat of the Noble strain (Noble et al., 1975), which is dependent on lactogenic hormones for prolif- eration (Gout, 1980; Richards et al., 1982). Functional analysis of different forms of the PRLR in a transient promotor/reporter assay system indicates that unlike the short form of the recep- tor, both the long form and the Nb2 forms are fully capable of transmitting PRES signal (Lesuer et al., 1991; Ali et al., 1992).

The mechanism by which these receptors transmit this sig- nal is not clear. However, it has become clear for receptors that do not possess any intrinsic tyrosine kinase catalytic domain that following stimulation by the ligand, there is an association with cellular tyrosine kinases. Recently, it has been shown for the EPOR (Witthuhn et al., 1993) and for the GHR (Argetsinger et al., 1993) that upon ligand stimulation these receptors asso- ciate with a nonreceptor tyrosine kinase of the J A K family (JAK2). Using the Nb2 cell line as a model system to study the action of PRL on cell mitogenesis, it has been reported that PRL stimulation results in tyrosine phosphorylation of a PRLR-as- sociated tyrosine kinase as well as of other cellular proteins (Rillema et al., 1992; Rui et al., 1992). Moreover it is clear that receptor dimerization and/or oligomerization is an important step in signal transduction for many hormones. Using the Nb2 cell line, we have demonstrated previously that monoclonal antibodies directed against the extracellular domain of the PRLR are capable of inducing mitogenesis (Elberg et al., 1990). Finally following PRL stimulation and tyrosine phosphoryla- tion of the associated kinase, little or nothing is known about the downstream events responsible for signal transduction. Some of the receptors of this family have been shown to be phosphorylated in response to ligand stimulation, such as EPOR (Yoshimura and Lodish, 19911, GHR (Foster et aZ., 1988; Carter-Su et al., 1989; Wang et al., 19931, interleukin-2P (Shackelford and “rowbridge, 1991) and gp130 (Murakami et al., 1991). In order to better understand the signal transduction process following kinase activation, we have examined phos- phorylation at the receptor level, and we clearly show for the first time that the PRLR itself is phosphorylated in Nb2 cells

14021

14022 PRL Receptor and JAK2 Phosphorylation

after hormone stimulation in a time- and dose-dependent man- ner. We also show that the associated kinase is JAK2, identical to the kinase activated by EPOR and GHR. Finally, we dem- onstrate that receptor homodimerization is also an important and necessary event for the phosphorylation of JAK2 and the receptor, as well as for the Nb2 cell proliferation.

EXPERIMENTAL PROCEDURES Materials-Pituitary hormones: human PRL was a generous gift

from Dr. S. Richards, Genzyme, Framingham, MA), ovine PRL (NIAMDD PRL-15) and ovine GH (NIH-GH-S11) were obtained from the National Hormone and Pituitary P r o g r a d I D D K (Baltimore, MD). Human GH (lot number 12404057) was kindly provided by Dr A. Ythier, Serono Aeres, Geneva, Switzerland. The Nb2 lactogen-dependent cell line (Nb2-11C) was developed and provided by Dr. P. Gout (Vancouver, Canada).

Cell Culture-Nb2-11C cells were normally grown in a complete me- dium (RPMI 1640 containing 10% fetal calf serum, 10% horse serum (Sigma category number H1895), 2 mM L-glutamine, 5 mM HEPES buffer, pH 7.4, penicillidstreptomycin (50 IU/ml and 50 pg/ml, respec- tively) and 50 of 2 P-mercaptoethanol @ME). Before stimulation, cells were incubated overnight a t 1.5 x lo6 celldml in a starvation medium (same as above without fetal calf serum and PME). Cell viabil- ity was assured before each experiment by counting cells after trypan blue coloration.

Cell Proliferation Assay-Nb2-11C cells were grown in complete me- dium and transferred to starvation medium overnight at a density of 2 x lo6 celVml. Cells were dispensed in 96-well plates at a density of 150,000 cells/l50 pl in starvation medium in the presence of oPRL (lo-$ M). Cells were stimulated for 24 h and then 1 pCi of [3Hlthymidine (Amersham TRA 120, 5 Cilmmol) was added in in 50 pl of starvation medium for a 4-h period of labeling. Cells were harvested using a Skatron microcell harvester (OSI, Inc., Paris) and counted in a P counter.

PRLR Labeling with Biotinylated mAb and Stimulation-For West- ern blot analyses, cells were pelleted and resuspended in starvation medium at a concentration of lo8 cells/ml, for 30 min a t 37 "C in the presence of 2 pg/ml ofmAb U5 directed against PRLR or of a n irrelevant mouse IgGl (MOPC 21, Sigma M9269), both coupled to biotin (Calbio- chem, catalog number 203-188), according to the manufacturer's in- structions. Cells were stimulated with the different hormones at lo" M (oPRL, oGH, hPRL, hGH) or hGH coupled to biotin. For the experiments with a tyrosine kinase inhibitor, the cells were incubated 18 h in pres- ence of herbimycin A (a generous gift from Dr. Uehara, Tokyo, Japan), a t a concentration of 0.1 pg/ml in the absence or presence of PME (for inactivation), prior to stimulation by lactogenic hormones.

Cross-linking with DSP-After stimulation for different times and at different concentrations depending on the experiment, cells were quickly pelleted in an Eppendorf centrifuge and washed once with 1 ml of phosphate-buffered saline containing 1 mM orthovanadate and 1 mM MgCl,. Cells were pelleted again and resuspended in 1 ml of the same buffer containing 0.5 mM of the thiol-cleavable cross-linker DSP (Pierce Chemical Co.) and incubated on ice for 20 min. The reaction was quenched by adding 10 pl of 1 M Tris-HC1, pH 7.5, and after centrifu- gation, the cell pellets were frozen in liquid nitrogen.

h at 4 "C in 1 ml of lysis buffer (10 mM Tris-HC1, pH 7.5; 5 mM EDTA, 50 Immunoprecipitation of Proteins-Cell pellets were solubilized for 1

mM NaCl, 30 mM sodium pyrophosphate, 50 mM sodium fluoride, 100 PM sodium orthovanadate, 0.2% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 5 pg/ml aprotinin, 1 pg/ml pepstatin A, 2 pg/ml leupeptin). Insoluble material was removed by 15 min of centrifugation at 15,000 x g, at 4 "C, and the supernatants were incubated overnight a t 4 "C, with a monoclonal antibody against PRLR (U6) a t a concentration of 5 pdrnl. The complexes were precipitated with anti-mouse IgG-agarose beads for 1 h a t 4 "C. After three washes in 1 ml of lysis buffer, precipitated material was eluted from the beads by boiling in 15 pl of elution buffer (20% glycerol; 10% PME; 4.6% SDS; 0.125 M Tris, pH 6.81, subjected to 7.5% SDS-PAGE under reducing conditions, and then transferred onto a polyvinylidene difluoride membrane (Immobilion, Millipore, catalog number lPVH 00010) for 45 min a t room temperature at a fixed current of 119 mA. The filters were blocked in 200 mM NaC1,50 m Tris, pH 7.5, 1% bovine serum albumin, 0.5% azide for 1 h at room temperature or overnight at 4 "C.

Precipitation with Avidin-agarose Beads-For the experiments using cell surface biotinylation of PRLR, the pellets were solubilized in the same lysis buffer as described above for 1 h a t 4 "C under rotation. The insoluble material was also removed by centrifugation a t 15,000 x g for

15 min and the supernatant incubated with 50 pl of avidin-agarose beads (lo%, PN) in lysis buffer (Calbiochem reference number 189742) for 1 h at room temperature under agitation. After three washes with 1 ml of washing solution (described before), proteins were eluted in 12 pl of elution buffer (see above). The samples were boiled for 10 min before loading onto a 7.5% SDS-PAGE under reducing conditions. After trans- fer, the blots were blocked as described before.

Immunoblot Analyses-After blocking, the filters were incubated for 90 min a t room temperature with the mouse monoclonal anti-phospho- tyrosine antibody (UBI catalog number 05321) at a concentration of 0.2 pg/ml in antibody buffer (50 mM Tris, pH 7.6, 200 mM NaCl, 0.05% Tween 20, 1% bovine serum albumin). The blots were washed twice for 5 min in washing buffer (50 m Tris-Cl, pH 7.6, 200 mM NaC1, 0.05% Tween 20), followed by incubation of 1 h at room temperature with either anti-mouse-IgG-HRP (Amersham NA 931) or anti-rabbit-IgG- HRP (Amersham NA934) diluted at 1/4000 in blocking buffer. After four washes of 30 min each, the blots were incubated with ECL chemilumi- nescence reagents according to the manufacturer's instructions (Amer- sham catalog number RPN 2106) for 1 min before exposure to x-ray film for 10-60 s.

Membrane Stripping and Rehybridization with Other Antibodies-In order to rehybridize the membranes with other antibodies, they were stripped overnight in an acid solution (0.1 M glycine, pH 3; 0.1 M NaCl) and washed with water before rehybridization at 4 "C for 2 hours either with a biotinylated monoclonal antibody against the PRLR (U5) at 0.5 pg/ml, or anti-JAK1 (UBI 06-254) or JAK2 (UBI 06-255) antibodies at a dilution of 1/2000. For the hybridization with the anti-JAK antibodies the membranes were then processed as described before. For the hy- bridization with U5, after washing the membranes were incubated 1 hour a t room temperature with horseradish peroxidase-conjugated streptavidin (Kierkegaard and Perry Laboratories, Inc., catalog number 143000) diluted a t 1/20,000. After four washes of 20 min each, ECL detection was performed as described above.

RESULTS

We have developed an improved detection technique consist- ing of specific labeling of the PRLR at the cell surface with a biotinylated monoclonal antireceptor antibody. As shown in Fig. L 4 , after the anti-phosphotyrosine immunoblot, the three lactogenic hormones (hGH, hPRL, and oPRL) induce tyrosyl phosphorylation of two proteins of 134 and 62 kDa, whereas oGH only stimulated weakly, if at all, the phosphorylation of the two proteins. This indicates that tyrosine phosphorylation of both proteins is specific to lactogenic hormones. In the ab- sence of cross-linker, the same results were obtained (data not shown), but with a decrease in intensity of the 134-kDa signal. The same study was performed by immunoprecipitation using a mAb (U6) against the PRLR (Fig. 1 D ) with similar results, except a very strong nonspecific signal at 55 kDa was seen, due to IgG heavy chains. In order to confirm if the smaller protein was the PRLR, we stripped the membrane and reprobed it with a monoclonal antibody directed against the extracellular por- tion of the PRLR (U5). A phosphorylated band of 62 kDa (Fig. lB) is recognized by the mAb, indicating that this band is indeed the PRLR. Moreover this band is equally expressed in unstimulated and stimulated cells, confirming that the same amount of protein was loaded onto the gel. Aweak band at 124 kDa is also seen, which may represent a dimeric form of the Nb2 PRLR. To ensure the specificity of our results, we reprobed the same membrane, after stripping, with another monoclonal anti-PRL (U6), directed against a different epitope in the ex- tracellular domain of the PRLR. The results (data not shown) are identical to those obtained with U5, demonstrating that the 62-kDa band is the PRLR. Because PRLRs on the cell surface were labeled with a biotinylated antibody, the heavy chains of this mAb generate the faint band observed at 55 kDa, present in all the samples, in Fig. lA. The recent observations of the involvement of JAK2 as the tyrosine kinase associated with the EPOR (Witthuhn et al., 1993) and the GHR (Argetsinger et al., 1993) and its theoretical molecular weight of 130,000 (Silven- noinen et al., 1993) prompted us to use an anti-JAK2 antibody

A Mr

200-

116-

80 -

50 -

1 2 3 4 5

c ”

c

c I

P R L Receptor and JAK2 Phosphoqdation

B Mr Mr

200 -

t 134

116-

c 62 80 -

50-

D 1 2 3 4 5

- 134

t 62

1 2 3 4 5

””-

O U ”

M,

t 124

t 62

c 1 2 3 4 5

E 1 2 3 4

e 55 Biotinylated control IgC: + +

Biotinylated anti PRLR: - + +

e 22 + o P R L : - + - +

t- 134

t 62

1402.7

\I

- 134

t 5;

FIG. 1. Q r o s i n e phosphorylation of thc. I ’ R I , R and its associated JAK kinase. After specific crll surfncr Iahrling of thr PRI,R wlth n hiotinylated mAh (Us), NIL! wlls w v w stilnul:ltrd \I I th lo”,’ hl of the lactogenic hormones, oPRL. h(;H. hPRI, (1nnr.s 2. 4 . 5 1. nnnlnrto~rnir d ; f I Ilnnr 3). or nonstimulated tlnnr, I I . After chemical crnss-llnkinF: with the DSI’, the cells wrre soluhllizrd and thr protrlns pur i f id t v l t h :tvldin-;lmrnsr heads before srparation hy SDS-PAGE and analysis by immunohlot with different antihodirs. A , ;Inti-phn4phot?.rrlElnl. ~ m m ~ ~ n r ~ h l r ~ t : 13. nnt~-I’Rl,R immunohlot; C, anti-JAK2 immunohlot. D, the Nh2 cells wrrr stimulatrd hy the diffmrnt llormonrs at I O v. ~IPRI,, o(;fl, h ( ; l l , and hl’ l~l ,~lrrrr~~u 2, 3. 4, and .5, respectively), or nonstimulated (Innr I ) , soluhilized, and the proteins werr immunoprrcipitntrd \ v l t h an ;~nti-l’l<l,R 1 [ . G I a t 5 ~ ~ g ’ m l . overnight, hcforc separation on SDS-PAGE and Western analysis with an anti-phosphotyrnsine ;mt lh ly . E. SI12 crIIq wrrv Inlwlrrl wlth rlthrr hiotinvlnted UT, ( Innrs 3 and 4 ) or a hiotinvlated control I& (Innrs I and 2 ). then stimulated with oPRI. I Innrs P and 4 I or not 1 I n n r ~ I :Ind .’I f. hrfnrr Westrrn analysis with an anti-phosphotyrosine antibody.

in an immunohlot analysis. Fig. 1C shows that the 134-kDa hand is recognized by the anti-.JAK2 antibody, suggesting that the kinase involved in the signaling pathway of the PRLR is prohahly JAK2 itself, since anti-dAK1 antibody failed to recog- nize this protein. JAK2 seems to he constitutively associated with the receptor, since no significant difference in the amount of J A W protein between the control and the stimulated samples was seen in three independent experiments (Fig. 1C and data not shown). Finally the specificity of the assay has been verified using an irrelevant hiotinylated control I g G for the “in oivo” laheling step, which failed to give any signal after a-phosphotyrosine immunoblot, as shown in Fig. 1E.

In order to confirm that phosphorylation of the PRLR was on tyrosyl residues, we incubated Nh2 cells with a tyrosine kinase inhibitor, herbimycin A. As shown in Fig. 2 . 4 , in the presence of the inhibitor, phosphorylation of the kinase and the receptor is completely abolished, but this effect can be reversed by coincu- hation with a reducing agent (BME), which inactivates the inhibitor (Ueharaet nl . , 1989). Equal amounts ofNh2 PRLRare expressed, independent of the presence of inhibitor (Fig. 2R).

This indicates that the decrease in tyrosine phosphorylation ohserved for both .JAW kinase and the PRLR is not dur to a toxic effect of herhimycin on the cells, hut rather to inhihition of a tyrosine kinase activity.

To determine the time course of the phosphoqlntinn of the receptor itself, we stimulated S h 2 cells for diffrrrnt prriods with oPRL (lo-‘ v ) following cell surface laheling uith hiotiny- Iated UFi. After chemical cross-linking and lysis, the rrcrptnr complexes wrre purified with avidin-agarose heads. In Fig. 31. maximum activation of JAK2 is seen at 5 min and of t h r re- ceptor at 20 min, indicating that phosphorylation of t h r recep- tor is delayed and requires . J A U phosphorylation. Fig. .?!I shows the same hlot after stripping and rrhyhridization with monoclonal antihody U.5 directed against the PRLR. indicatinc that the differences, ohservrd in phosphorylation (Fig. 3.4 1, r l n not simply represent a difference of receptor expression. These results suggest that the PRLR could he a substrate fnr r i thrr JAK2 or an intermediate transducrr molecule. activated within 5-20 min.

In order to vrrifv that phosphorylation of PRLR occurs a t

14024 P R L Receptor and JAK2 Phosphordotion

I" -

I- Biotinylated hGH :

Herbimycin A :

Reducing agent :

+ + + - + +

+

1 2 3

' I

Mr

c- 134

t- 62

- 124 - 62 Biotinylated hGH: + + +

Herbimycin A : - + + Reducing agent : - - +

Ftc:. 2 . Tyrosine kinase inhihition of the phosphorylation of JAW and the PItLR. Nh'L cells wrrr starvrd overnight in thr prrsrncr (lotrcs 2 antl 3 I or :tl>senrr tlnrrr 1 ) of t h c tyrosinv kinasr inhihitor hrrhimycin A (0.1 pg/mI), plus [$ME (50 p v ) ( lnnc 3 1 . Aftrr 10 min of stimulation w i t h 10 ',' L! of hiotinyl:ltrd hGFI. crlls wrre prncrssc4 as drscrihrd in thr Irgrnd to Fig. 1 and thr protrins annlvzrtl hv immu- nohlot w i t h an anti-phosphotvrosinr antihotlv at 0.2 pgiml ( A I or anti- Pl<I,Ii mAh : I t 0.5 pg/ml ( f j ).

physiological concentrations, we incuhated Nh2 cells with in- creasing concentrations of oPRL. As shown in Fig. 4C, after immunoprecipitation with an anti-PRLR antihody (U6) and Western analysis with an anti-phosphotyrosine antihody. phos- phorylation of hoth the kinase and PRLR can he seen at con- centrations as low a s IO-'' 11 and increase to reach a plateau at lo-' M . An immunohlot using a monoclonal antihody (U5) re- veals no difference in the c.xpression of the receptor (Fig. 4D). The maximum response is ohserved at higher concentrations of hormone than has been ohserved previouslv Rui rt a / . , 1992 J, which could he due to the fact that cells were stimulated for 20 min (maximal response for receptor), which is much longer than the optimal time for stimulation ofdAK2 phosphorylation.

' 1 2 3 4 5 6 7 8

Mr

t- 134

e- 62

Mr

- 124 - 62

Intrrrstingly, at concrntrations highrr than 10" \I. phnspho- rylation of .JAK2 and tht. rrceptnr tlcclinc*. A simil:lr dnw-rr- sponstl analysis was ohsrnrd af t r r crII .surface Iahrlinr! o f thv PRLR with a hiotinylatetl mAh (L'5). as shotvn in Fig. .I.,\. Fig. 4f3 rrprrsents the \\ 'rstrrn analysis of thr rwrptor. in which t.hc. same mrmhranr was stt-ippd and rrprohcd tvith m:Zh t -5 , A hand is ohservcd also at 55 kDa. which reprrsrnt.q thc hrzvy chain of t h r mAh used for t h r crll surfarr Iahrling.

Finally tve analyzed the effrct of incrrnsing concrnt rat ions o f Iactogrnic and nonlactogcnic hormonrs on Sh2 cell prnlifrra- t,ion, representing the final wvnt in t h r PRL s ipa l lng pn th- way. As shown in Fig. 5, all Iactogrnic hormonrs t w t r t l I n l ' R I , and hGM) hind thr PRLR with thr samr affinity and intlucr crll mitogenesis antl prolifrration at concrntr:ttions ns low :IS 1 0 L'

>I . A hiphasic CUT\'(' is s w n with incrrasing concc~ntr:Itionr reaching a platrau at IO"" \I and thr rrsponsr drclining at thr. highest concentrations (from 10.-'to 10 ' > ! I . Ovinv G I 1 was ahlr to inducr a similar hiphnsic cumr, hut 100-fold morr d ; F I is required. This is prohnh1.v tlur to contamination o f t h r nGiT hy oPRL, since recnmhinant hGf1 had no stimulating c.ffwt. Thc amount. of lactogenic hormone requirrd to stimulntr crll prnlif- cration is much lowrr thnn that nerdrd to activatr phnspho- rylation. This could he d u r to t h r rrlativc. short stimrrlntion time for phosphorylation (10 min), comparrd \vith thr 24-h exposure to hormone for t h r crll prolifrration.

I)IS~'I'SSIOT

For powth factor rwrptors that have an intrinsic k i n a w domain, it is known that follo\ving ligand stirnulation. transphosphorylation of the kinnsr occurs and thr phosphoty- rosy1 residues hind to SH2 domains of rffwtor rnolreulc~s I r : l d -

ing to activation. Krcrptors of t h r cytokinrl(;FlTKI, f>rmily do

A 1

I

-

C ,

PRL Receptor and JAK2 Phosphorylation

2 3 4 5 6 7 8

-

2 3 4 5 6 7 8

Mr

c- 134

C- 62

1 2 3 4 5 6 7 8

a

14025

Mr

- 124

c- 62 - 55

e cs -

C , L

Ilormone Concentration (M)

FK. 5. Effec t of i n c r e a s i n g c o n c e n t r a t i o n s of hormone on [ 'Hlthymidine incorporat ion of Nh2 cel ls . ARrr an ovcmight stnr- vntion, Nh2 crlls wrrr incuhntcd with tliffrrrnt concrntrations I O - ' ! to 10~'a~)ofol 'RI,orhl 'KI,! ' ' l , n ( ; I H ( ' ~ , h ( ~ I I ~ O ~ , o r r r c o m h i n n n t h G H ~ ~ ~ for 24 h hrforc, nddition of 1 ~ C I of I 'H/thym1dinr for 4 h. Thc crlls w(1r.r hnrvcstrd, Ivsrtl. :1nd thr incorporntrtl radioactivity countrd.

not have any intrinsic catalytic domain, hut have heen shown to associate with a tyrosine kinase that results in the phospho- rylation of th i s k inase and in some cases of the receptor itself.

I t is known that erythropoirt in induccs thr tyrosinr phospho- rvlation o f the kinase and of the carhnxyl-trrminal portion of the EPOR tYnshimura and Lodish. 1991 1. Similarly. thr ( ; l iR has heen reported to he phosphoFlatcd along \vith i t s nssnri- atrd kinase (Fostrrc7t nl.. 19SH: Car tw-Su ( 3 1 n l . , 1 W 9 ; \V:lnc v t nl., 19931. Rui r t ol. (19921 rcportcd t h c \vcak phosphoryl:ltion o f a 66-kDa hand, follmving immunoprrcipitntion tvith rithrr an anti-phosphotyrosine antihody or an anti-PRLR antihody (U6). that thry suggested \vas t he PRI, rrcrptor in S h 2 crlls. Sincr these cells arr in fact immunc crlls capnhlc o f .srrrrt inf IgG, and s ince the PRLR has a molccular mass of 62 k D a . not too different from thc size of thc hcavy chain of lgc;. i t is dificult to clcarly discern receptor phosphorylntion usinK t h r classical immunoprecipitation techniquc. In contrast, thfn prr- sent detection system is hasrd on thr spwific Iahrling of t hc molcnllr of interest a t t h c cell surfacr \vith a hiotinylatcd nn- t ihody directed against the cxtracellulnr domain of thcb rccrp- tor. \\'e clearly demonstrate that . fnllo\ving ligand stimul, < I t ' ton. a 62-kDn protein is strongly tyrosine-phosphory1ntc.d :rnd that this phosphoprotein cnrrcsponds to thc l'?<t,R in Sh? w l 1 . s . I\

second protein, nssociatrd with the PRLR. \vith :I sin' of 13.4 kDa, is also strongly tyrosine phosphor?.latrd in rrspnnsr to lactogenic hormones and corrrsponds to the kinasf,. Prr.viously. a protein of 120 kDa has heen rrportrd to hc phosphoryl:~tc~rl in response to PRL stimulation of Nh2 crlls (Rui r * t ol.. 1992: Rillema a/., 1992). %lore recently, the GHR nssoriatrd kinase has heen identified as .JAW (Argctsingcr 6.t r r l . . l W : < j , Xvhich h a s a theoretical molecular mass of 1.30 kDa (Si lvrnnoinrn vt nl.. 1993~, and was also drscrihctl to hc thc kinasr assori:ttrd with the EPOR (U'i t thuhn vt nl., 19931.

Our resul ts indicatr c lear ly that thr : lssociatr kin:1w in- volved in PRLR simal transduction is also a m r m h r r of thc, ,JAK family and. most. prohahly, .JAK2 i tsrlf . Thr ( I : I ~ : I sucgrqt t ha t t he k inase is constituvel.v aswciatcd with cbithrr the. rv- ceptor, or an in te rmedia te t ransducr r . and apprnrs t o h f a m p -

14026 PRL Receptor and JAK2 Phosphorylation

idly phosphorylated only after hormone stimulation, as it has been shown for the EPOR (Witthuhn et al., 1993). Ligand bind- ing results in the activation of the constituvely associated ki- nase, which following autophosphorylation, phosphorylates the PRLR itself or an intermediate protein kinase which is respon- sible for phosphorylation of the receptor. Tyrosine phospho- rylation of the Nb2 PRLR could be involved in binding to SH2 domains of other transducer molecules and lead to signal trans- duction and cell proliferation.

Ligand binding for some members of the family has been demonstrated to induce homodimerization. This is the case for the EPOR (Watowich et al., 1992) and GHR (Cunningham et a[., 1991). In fact the three-dimensional crystal structure of the extracellular binding domain of the GH receptor and hGH is the only definitive proof of such homodimerization (de Vos et al., 1992). Recently, it has been suggested that gp130 must dimerize to induce tyrosine phosphorylation in response to in- terleukin-6 (Murakami et al., 1993).

Several mechanisms of dimerization are possible, as have been described previously for receptors with tyrosine kinase activity (Ullrich and Schlessinger, 1990). A first subclass is induced by monovalent ligand, such as epidermal growth fac- tor, in which two molecules of ligand bind to two molecules of receptor inducing a conformational change of the receptor and dimerization. It is also possible for a monomeric ligand to in- teract with a disulfide-linked receptor dimer, as in the case of insulin, although it is still not clear if one or two molecules of ligand interact. Finally, the ligand itself can be responsible for dimerization. This is the case for hGH and its receptor, with two sites (1 and 2) on the ligand that interact sequentially to first one and then a second molecule of receptor (Fuh et al., 1992). It was also suggested that the human PRLR was able to dimerize after hormone stimulation, using a binding site 2 mutant of hGH (GlZOR), which is inactive in stimulating growth of Nb2 cells and can antagonize the effects of hGH or hPRL (Fuh et al., 1993). In the present study, using higher concentrations of lactogenic hormone (up to M), we ob- served a bell-shaped curve for Nb2 cell proliferation, but also a reduced activation of both JAK2 and PRLR phosphorylation, suggesting that dimerization of PRLR is an important and necessary step for signal transduction as it has been shown for the two-site model of GH action. There are other interpreta- tions of a bell-shaped curve, as desensitization or down-regu- lation of receptor. However, as seen in Fig. 4 B , the same level of receptor expression is seen at all concentrations of PRL, sug- gesting that down-regulation did not occur. In addition we have shown recently that dimerization of the PRLR in Nb2 cells is necessary for activation of the associated kinase and for cell proliferation, using monovalent Fab fragments.2 Finally, a non- reduced dimer form of the receptor appears to be present, al- though in very low amounts, which could explain why the anti- phosphotyrosine antibody failed to recognize this dimeric form.

These results, when considered together, strongly suggest that following binding of PRL to the Nb2 PRLR, dimerization of the receptor occurs prior to phosphorylation of an associated tyrosine kinase ( J B ) and mitogenesis. It will be interesting

H. Rui, J. J. Lebrun, R. A. Kirken, P. A. Kelly, and W. L. Farrar, submitted for publication.

to investigate the role of the phosphorylated tyrosines of both JAK2 and the receptor in the mitogenic response of these cells.

Acknowledgments-We thank Dr. Y. Uehara for providing the anti- biotic herbimycin A, Dr. S. Richards for the generous gifi of hPRL, and Dr. A. Ythier for kindly providing hGH.

Note Added in Proof-A paper recently appeared reporting that in Nb2 cells JAK2 is associated with the PRL receptor and activated by

5364-5368. PRL (Rui, H., Kirken, R. A,, and Farrar, W. L. (1994) J. Biol. Chem. 269,

REFERENCES

Ali, S., Edery, M., Pellegrini, I., Lesuer, L., Paly, J., Djiane, J., and Kelly, P. A. Mi, S., Pellegrini, I., and Kelly, P. A. (1991) J . Biol. Chem. 266, 20110-20117

(1992) Mol. Endocrinol. 6, 1242-1248 Argetsinger, L. S., Campbell, G. S., Yang, X., Witthuhn, B. A,, Silvennoinen, O.,

Ihle, J. N., and Carter-Su, C. (1993) Cell 74,237-244 Boutin, J. M., Jolicoeur, C., Okamura, H., Gagnon, J., Edery, M., Shirota, M.,

Banville, D., Dusanter-Fourt, I., Djiane, J., and Kelly, P. A. (1988) Cell 53,69-77 Carter-Su, C., Stubbart, J. R., Wang, X., Stred, S. E.,Argetsinger, L. S., and Shafer,

J. A. (1989) J. Biol. Chem. 264, 18654-18661 Comsa, J., Leonhardt, H., and Wekerle, H. (1982) Reu. Physiol. Biochem. Pharma-

col. 92, 115-191 Cosman, D. (1993) Cytokine 6,95-106 Cunningham, B. C., Ultsch, M., De Vas, A. M., Mulkerrin, M. G., Clauser, K. R., and

de Vos, A. M., Ultsch, M., and Kossiakoff, A. A. (1992) Science 266, 306312 Elberg, G., Kelly, P. A., Djiane, J., Binder, L., and Gertler, A. (1990) J. Biol. Chem.

266,14770-14776 Foster, C. M., Shafer, J. A., Rozsa, F. W., Wang, X., Lewis, S. D., Renken, D. A,,

Natale, J. E., Schwartz, J., and Carter-Su, C. (1988) Am. Chem. SOC. 27, 3 2 6 334

Fuh, G., Cunningham, B. C., Fukunaga, R., Nagata, S., Goeddel, D. V., and Wells,

Fuh, G., Colosi, P., Wood, W. I., and Wells, J. A. (1993) J. Biol. Chem. 288, 5376- J. A. (1992) Science 266, 1677-1680

5381 Gout, P. W., Beer, C. T., and Noble, R. L. (1980) Cancer Res. 40,2433-2436 Kelly, P. A., Ali, S., Roaakis-Adcock, M., Goujon, L., Nagano, M., Pellegrini, I.,

Gould, D., Djiane, J., Edery, M., Finidory, J., and Postel-Vinay, M. C. (1993)

Lesuer, L., Edery, M., Mi, S., Paly, J., Kelly, P. A,, and Djiane, J. (1991) Proc. Natl. Recent Frog. Horm. Res. 48

Murakami, M., Narazaki, M., Hibi, M.,Yawata, H.,Yasukawa, K., Hamaguchi, M., Acad. Sei. U. S. A. 88, 824-828

Taga, T., and Kishimoto, T. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 11349- 11353

Murakami, M., Hibi, M., Nakagawa, N., Nakagawa, T., Yasukawa, K., Yamanishi, K., Taga, T., and Kishimoto, T. (1993) Science 260, 1808-1810

Noble, R. L., Clayton-Hochachka, B., and King, D. (1975) Cancer Res. 36,766780 Pellegrini, I., Lebrun, J.J., Ali, S., and Kelly, P. A. (1992) Mol. Endocrinol. 6,

Richards, J. F., Beer, C. T., Bourgeault, C., Chen, K, and Gout, P. W. (1982) Mol. 1023-1031

Rillema, J. A., Cambell, G. S., Lawson, D. M., and Carter-Su, C. (1992) Endocri- Cell. Endocrinol. 26, 41-49

Rui, H., Djeu, J. Y., Evans, G. A., Kelly, P. A., and Farrar, W. L. (1992) J. Bid. Chem. nology 131,973-975

267,24076-24081 Russell, D. H., Matrisian, L., Kibler, R., Larson, D. F., Poulos, B., and Magun, B. E.

Sabhawal, P., Glaser, R., Lafuse, W., Varma, S., Liu, Q., Arkins, S., Kooijman, R., (1984) Biochem. Biophys. Res. Commun. 121,899-906

Kutz, L., Kelley, K. W., and Malarkey, W. B. (1992)Proc. Nafl. Acad. Sci. U. S. A. 89,7713-7716

Wells, J. A. (1991) Science 25, 821-825

Shackelford, D. A,, and Trowbridge, I. S. (1991) Cell Regul. 2 , 7 3 4 5 Shirota, M., Banville, D., Ali, S., Jolicoeur, C., Boutin, J.-M., Edery, M., Djiane, J.,

Silvennoinen, O., Witthuhn, B. A,, Quelle, F. W., Cleveland, J. L., Yi, T., and Ihle,

Uehara, Y., Fukazawa, H., Murakami, Y., and Mizuno, S. (1989) Biochem. Biophys.

Ullrich, A., and Schlessinger, J. (1990) Cell 61,203-212 Wang, X., Moller, C., Norstedt, G., and Carter-Su, C. (1993) J. Bid. Chem. 268,

Watowich, S. S., Yoshimura, A,, Lonmore, G. D., Hilton, D. J., Yoshimura, Y., and

Witthuhn, B. A,, Quelle, F. W., Silvennoinen, O., Yi, T., Tang, B., Miura, O., and

Yoshimura, A,, and Lodish, H. F. (1992) Mol. Cell. Biol. 12, 706-715

and Kelly, P. A. (1990) Mol. Endocrinol. 4,1136-1143

J. A. (1993) Proc. Natl. Acad. Sci. U. S. A. 90,842943433

Res. Commun. 163,803409

3573-3579

Lodish, H. F. (1992) Proc. Natl. Acad. Sci. U. S. A. 89,214&2144

Ihle, J . N. (1993) Cell 74, 227-236